Peptide-mediated delivery of therapeutic mRNA in ovarian cancer

Abstract

Ovarian cancer is the most lethal gynecological malignancy in the developed world. In spite of intensive research, the mortality has hardly decreased over the past twenty years. This necessitates the exploration of novel therapeutic modalities. Transient protein expression through delivery of mRNA is emerging as a highly promising option. In contrast to gene therapy there is no risk of integration into the genome. Here, we explore the expression of mRNA in models of ovarian cancer of increasing complexity. The cell-penetrating peptide (CPP) PepFect 14 (PF14) was used to formulate CPP-mRNA nanoparticles. Efficient expression of a reporter protein was achieved in two-dimensional tissue cultures and in three-dimensional cancer cell spheroids. PF14 nanoparticles greatly outperformed a lipid-based transfection agent in vivo, leading to expression in various cell types of tumor associated tissue. Protein expression was restricted to the peritoneal cavity. Messenger RNA expression across different cell types was confirmed in primary ovarian cancer explants. As ovarian cancer is confined to the peritoneal cavity in most cases, the results create the basis for applications in which the tumor microenvironment is transiently modified through protein expression.

Introduction

Epithelial ovarian cancer (EOC) is the most lethal gynecological malignancy in the developed world, and the sixth leading cause of cancer mortality in women [1]. Since the tumor readily distributes throughout the abdominal cavity without causing any symptoms, it is often only discovered when widespread metastases are present. In an advanced stage, EOC therapy consists of cytoreductive surgery and adjuvant chemotherapy in order to remove as much of the tumor as possible. However, (micro)metastases persevere and tumor recurrence is almost inevitable. Median progression-free survival (PFS) of EOC is only 18 months, and resistance to chemotherapy is common after tumor recurrence [2].

Emerging novel therapies are the tyrosine kinase inhibitor cediranib and the monoclonal antibody bevacizumab that both inhibit angiogenesis. Cediranib significantly increased PFS in combination with chemotherapy [3]. Combination therapy of bevacizumab with chemotherapy also showed increased PFS in newly diagnosed and platinum resistant ovarian cancer [4], [5]. It furthermore showed an increase in overall survival in a subcategory of poor-prognosis patients [5]. Poly(ADP-ribose) polymerase (PARP) inhibitors are another class of drugs that are currently under investigation and olaparib showed increased PFS in platinum-sensitive cases [6]. Despite the increase in PFS, however, none of these therapies yield increased overall survival. As a consequence, novel therapeutic options are urgently needed.

Remarkably, extraperitoneal metastases of ovarian cancer are only observed in 12–33% of the patients diagnosed with EOC [7]. This trait is exploited with the use of intraperitoneal (IP) chemotherapy, in which a catheter through which chemotherapy can be applied is left after complete or optimal debulking surgery. This therapy has shown benefit for PFS. In pre-clinical research, IP delivery of nanomedicines has also been explored. This includes IP delivery of small interfering RNA (siRNA) [8], [9], [10], [11], [12], [13], [14], [15].

Next to downregulation of protein expression by RNA interference, protein induction by gene delivery has been studied for more than 25 years [16]. While this approach is mostly directed towards reconstitution of expression of defective genes, it has also been used for expression of proteins in cancer therapy. Examples regarding ovarian cancer include the induction of herpes simplex virus thymidine kinase (HSV-TK) and the restoration of P53 and BRCA-1 function [17], [18], [19].

One of the obstacles in DNA-based gene therapy is random insertion into the host genome which can induce carcinogenesis [20]. Furthermore, expression can last for uncontrolled periods of time, which is not desired in an acute intervention. Messenger RNA (mRNA) is emerging as a promising alternative that yields a transient protein expression [21]. The focus of this field is still in the area of vaccination, even though the feasibility of therapeutic protein expression, also in large animals, has been demonstrated and clinical trials are underway [22], [23], [24], [25].

As for any oligonucleotide-based therapy, the application of mRNA critically depends on efficient delivery vectors. These vectors have to protect the mRNA from enzymatic degradation and mediate efficient cellular uptake and endosomal release. In addition, they also need to shield the mRNA from recognition by the innate immune system [21], [26]. Various polymer and lipid-based delivery systems have been explored [26], [27]. In most cases, the oligonucleotides are packaged into nanoparticles through non-covalent, charge-driven interactions with a cationic or ionizable delivery agent. Following cellular uptake, induction of efficient endosomal escape is a distinctive characteristic of active delivery vectors [28].

For mRNA-based therapeutic protein expression the focus has been on lipid nanoparticles (LNPs) [29]. By comparison, the (pre-)clinical development of polymers for mRNA delivery is lagging behind. This may be attributed to the fact that an increase in uptake efficiency often correlates with an increase in toxicity as for example for polyethyleneimine [30]. Furthermore, either efficient breakdown and/or excretion have to be ensured. Cell-penetrating peptides (CPPs) constitute an interesting option that combines a low molecular weight with efficient proteolytic break-down. CPPs are a class of mostly cationic peptides of 8–30 amino acids in length that are able to induce cellular uptake of themselves and conjugated cargo [31]. They form nanoparticles with oligonucleotides through non-covalent interactions. Cellular uptake has little predictive power for cytosolic delivery of oligonucleotides, and only few CPPs yield efficient cytosolic oligonucleotide delivery [32]. One CPP with superior delivery capacity for antisense oligonucleotides, siRNA, and plasmid DNA (pDNA) in vitro and in vivo is PepFect 14 (PF14) [33], [34], [35], [36]. This cationic amphipathic CPP consists of 21 amino acids, of which 5 are positively charged, and comprises an N-terminal stearylation.

Regardless of the incorporation of targeting moieties, oligonucleotide polyplexes and LNPs show a strong propensity for liver targeting [12], [14], [37], [38], [39], [40]. Combining CPPs with targeting modalities such as peptides shows only limited success in changing the biodistribution [41]. Therefore, we propose an intraperitoneal approach, akin to intraperitoneal chemotherapy, for exploring the feasibility of mRNA delivery in ovarian cancer. As a basis for future therapy design, we wanted to learn which cell types would be reached in the heterogeneous context of a tumor in vivo. Based on the superior characteristics of PF14 for oligonucleotide delivery we chose this CPP for formation of mRNA nanoparticles. For comparison, we selected a commercial lipid-based transfection agent. Following the demonstration that both agents yielded mRNA-dependent protein expression in two-dimensional tissue cultures and three-dimensional tumor spheroids, we assessed protein expression in an intraperitoneal mouse model of ovarian cancer and in human tumor explants. For the murine model, nanoparticles were injected intraperitoneally.

PF14 nanoparticles targeted the tumor but not exclusively the tumor cells within the tumor. Reporter proteins were observed in fibroblasts, tumor cells, and immune cells. The lipid-based formulations did not show any uptake or translation in the xenograft model. Furthermore, there was no expression outside the abdominal cavity. As a consequence, we consider intraperitoneal mRNA delivery a highly interesting option to transiently modulate the peritoneal tumor microenvironment (TME).